Such network components are generally known. So-called switches or bridges analyze the data traffic in the network and then after the analysis take logical decisions in respect of the forwarding of incoming datagrams by way of either the first, second or third port. To this end known network components manage a so-called forwarding database (FDB), in which, in addition to an address, normally a so-called MAC address, at which the network component receives a frame or one or a number of datagrams included therein—hereafter referred to collectively as a telegram—from another communications subscriber, the physical port, in other words the first, second, third, etc. port, at which this was received, is also stored. If the network component later receives a telegram, which is sent to an address associated with a communications subscriber, from which a telegram has already been received, it is possible to determine, by checking the FDB, by way of which of the ports of the network component this communications subscriber can be reached, and the telegram is forwarded in a correspondingly specific manner by way of this port. If a destination address is not yet known, known network components route the relevant telegram to all active ports.
Disadvantages of these known network components are on the one hand the resources required to hold the FDB ready and on the other hand the unfavorable conditions that result when a destination address is not yet known and telegrams are sent m a correspondingly unnecessary manner into a number of segments of the network, with the result that unnecessary bandwidth is taken up and is not available for other purposes for the relevant time interval. Also there is no need for an often time-consuming search in an FDB or similar with delayed forwarding proving unavoidable even with comparatively efficient, so-called hash searches due to the inevitable scope of such databases. A delay resulting from such searches would also be incurred at each forwarding station, so the delay time quickly adds up to tangible orders of magnitude.
A method for configuring a control system of an electrical switching unit is known from US 2002/0059477 A1, wherein the switching unit comprises a primary device and a number of control devices, which are connected to one another by buses in the known manner, such that the primary device is connected to a higher-order bus and the control devices have an internal process bus. A machine control system is known from EP 1 184 754 A2, wherein the system is divided into different hierarchies and communications relations exist in each hierarchical level between the system components there.